Approximately half of all sexual assaults involve alcohol consumption by the perpetrator, victim, or both. The goal of the proposed research is to develop a virtual reality simulation that can be used as a proxy for perpetration of sexual aggression in alcohol administration studies. Self-report surveys conducted with victims and perpetrators have provided a wealth of important information about alcohol's role in sexual assault; however, causality cannot be established from correlational designs. In alcohol administration studies, participants can be randomly assigned to drink conditions, thereby allowing causal conclusions to be made. The major challenge for experimentalists is to develop proxies for sexual assault that have strong construct validity and experimental realism. Virtual reality environments provide the opportunity for participants to become immersed in the simulated environment; thus, participants are expected to behave in ways and to make choices that closely relate to their behavior outside the laboratory. The Specific Aims of the proposed research involve developing a virtual reality simulation with a virtual woman (an embodied female agent) that male participants rate as realistic and involving. The script of the interaction will follow that of a prototypic acquaintance sexual assault situation, allowing male participants to make choices about sexual activities in which they want to engage and (if they choose) to repeatedly try to make the female agent engage in sexual activity despite her protests. Most participants will be young, single, sexually active, heterosexual men to correspond to the demographics of most perpetrators. Focus groups and an iterative series of pilots that include cognitive interviews will be conducted with men and women to develop the simulation. Then a preliminary assessment of construct validity will be conducted by comparing self-reports of past sexual aggression, common risk factors, and responses to another proxy with participants' responses during the simulation. Also, an initial evaluation of the simulation's usefulness in alcohol administration research will be conducted by comparing the responses and actions of intoxicated and sober participants during the simulation. The findings from this study will serve as pilot data for a R01 proposal that will formally evaluate the paradigm's construct validity and usefulness in identifying men predisposed to commit sexual assault while intoxicated. Future studies can also alter various aspects of the simulation to increase generalizability to different populations. The long-term goal is to develop prevention and treatment interventions, thereby reducing rates of sexual aggression and addressing Healthy People 2020 goals.